Accelerated Gradient Methods for Stochastic Optimization and Online Learning

Size: px
Start display at page:

Download "Accelerated Gradient Methods for Stochastic Optimization and Online Learning"

Transcription

1 Acceleraed Gradien Mehods for Sochasic Opimizaion and Online Learning Chonghai Hu, James T. Kwok, Weike Pan Deparmen of Compuer Science and Engineering Hong Kong Universiy of Science and Technology Clear Waer Bay, Kowloon, Hong Kong Deparmen of Mahemaics, Zhejiang Universiy Hangzhou, China Absrac Regularized risk minimizaion ofen involves non-smooh opimizaion, eiher because of he loss funcion (e.g., hinge loss) or he regularizer (e.g., l -regularizer). Gradien mehods, hough highly scalable and easy o implemen, are known o converge slowly. In his paper, we develop a novel acceleraed gradien mehod for sochasic opimizaion while sill preserving heir compuaional simpliciy and scalabiliy. The proposed algorihm, called (Sochasic Acceleraed GradiEn), exhibis fas convergence raes on sochasic composie opimizaion wih convex or srongly convex objecives. Experimenal resuls show ha is faser han recen (sub)gradien mehods including, and. Moreover, can also be exended for online learning, resuling in a simple algorihm bu wih he bes regre bounds currenly known for hese problems. Inroducion Risk minimizaion is a he hear of many machine learning algorihms. Given a class of models parameerized by w and a loss funcion l(, ), he goal is o minimize E XY [l(w;x,y )] w.r.. w, where he expecaion is over he join disribuion of inpu X and oupu Y. However, since he join disribuion is ypically unknown in pracice, a surrogae problem is o replace he expecaion by is empirical average on a raining sample {(x,y ),...,(x m,y m )}. Moreover, a regularizer Ω( ) is ofen added for well-posedness. This leads o he minimizaion of he regularized risk min w m m l(w;x i,y i ) + λω(w), () i= where λ is a regularizaion parameer. In opimizaion erminology, he deerminisic opimizaion problem in () can be considered as a sample average approximaion (SAA) of he corresponding sochasic opimizaion problem: min w E XY [l(w;x,y )] + λω(w). () Since boh l(, ) and Ω( ) are ypically convex, () is a convex opimizaion problem which can be convenienly solved even wih sandard off-he-shelf opimizaion packages. However, wih he proliferaion of daa-inensive applicaions in he ex and web domains, daa ses wih millions or rillions of samples are nowadays no uncommon. Hence, off-he-shelf opimizaion solvers are oo slow o be used. Indeed, even ailor-made sofwares for specific models, such as he sequenial minimizaion opimizaion (SMO) mehod for he SVM, have superlinear compuaional

2 complexiies and hus are no feasible for large daa ses. In ligh of his, he use of sochasic mehods have recenly drawn a lo of ineres and many of hese are highly successful. Mos are based on (varians of) he sochasic gradien descen (SGD). Examples include Pegasos [], SGD-QN [], [3], and sochasic coordinae descen () [4]. The main advanages of hese mehods are ha hey are simple o implemen, have low per-ieraion complexiy, and can scale up o large daa ses. Their runime is independen of, or even decrease wih, he number of raining samples [5, 6]. On he oher hand, because of heir simpliciy, hese mehods have a slow convergence rae, and hus may require a large number of ieraions. While sandard gradien schemes have a slow convergence rae, hey can ofen be acceleraed. This sems from he pioneering work of Neserov in 983 [7], which is a deerminisic algorihm for smooh opimizaion. Recenly, i is also exended for composie opimizaion, where he objecive has a smooh componen and a non-smooh componen [8, 9]. This is paricularly relevan o machine learning since he loss l and regularizer Ω in () may be non-smooh. Examples include loss funcions such as he commonly-used hinge loss used in he SVM, and regularizers such as he popular l penaly in Lasso [], and basis pursui. These acceleraed gradien mehods have also been successfully applied in he opimizaion problems of muliple kernel learning [] and race norm minimizaion []. Very recenly, Lan [3] made an iniial aemp o furher exend his for sochasic composie opimizaion, and obained he convergence rae of O ( L/N + (M + σ)/ N ). (3) Here, N is he number of ieraions performed by he algorihm, L is he Lipschiz parameer of he gradien of he smooh erm in he objecive, M is he Lipschiz parameer of he nonsmooh erm, and σ is he variance of he sochasic subgradien. Moreover, noe ha he firs erm of (3) is relaed o he smooh componen in he objecive while he second erm is relaed o he nonsmooh componen. Complexiy resuls [4, 3] show ha (3) is he opimal convergence rae for any ieraive algorihm solving sochasic (general) convex composie opimizaion. However, as poined ou in [5], a very useful propery ha can improve he convergence raes in machine learning opimizaion problems is srong convexiy. For example, () can be srongly convex eiher because of he srong convexiy of l (e.g., log loss, square loss) or Ω (e.g., l regularizaion). On he oher hand, [3] is more ineresed in general convex opimizaion problems and so srong convexiy is no uilized. Moreover, hough heoreically ineresing, [3] may be of limied pracical use as () he sepsize in is updae rule depends on he ofen unknown σ; and () he number of ieraions performed by he algorihm has o be fixed in advance. Inspired by he successes of Neserov s mehod, we develop in his paper a novel acceleraed subgradien ( scheme for sochasic composie opimizaion. I achieves he opimal convergence rae of O L/N + σ/ ) N for general convex objecives, and O ( (L + µ)/n + σµ /N ) for µ- srongly convex objecives. Moreover, is per-ieraion complexiy is almos as low as ha for sandard (sub)gradien mehods. Finally, we also exend he acceleraed gradien scheme o online learning. We obain O( N) regre for general convex problems and O(log N) regre for srongly convex problems, which are he bes regre bounds currenly known for hese problems. Seing and Mahemaical Background Firs, we recapiulae a few noions in convex analysis. (Lipschiz coninuiy) A funcion f(x) is L-Lipschiz if f(x) f(y) L x y. Lemma. [4] The gradien of a differeniable funcion f(x) is Lipschiz coninuous wih Lipschiz parameer L if, for any x and y, f(y) f(x) + f(x),y x + L x y. (4) (Srong convexiy) A funcion φ(x) is µ-srongly convex if φ(y) φ(x)+ g(x),y x + µ y x for any x,y and subgradien g(x) φ(x). Lemma. [4] Le φ(x) be µ-srongly convex and x = arg min x φ(x). Then, for any x, φ(x) φ(x ) + µ x x. (5)

3 We consider he following sochasic convex sochasic opimizaion problem, wih a composie objecive funcion min{φ(x) E[F(x,ξ)] + ψ(x)}, (6) x where ξ is a random vecor, f(x) E[F(x,ξ)] is convex and differeniable, and ψ(x) is convex bu non-smooh. Clearly, his includes he opimizaion problem (). Moreover, we assume ha he gradien of f(x) is L-Lipschiz and φ(x) is µ-srongly convex (wih µ ). Noe ha when φ(x) is smooh (ψ(x) = ), µ lower bounds he smalles eigenvalue of is Hessian. Recall ha in smooh opimizaion, he gradien updae x + = x λ f(x ) on a funcion f(x) can be seen as proximal regularizaion of he linearized f a he curren ierae x [6]. In oher words, x + = arg min x ( f(x ),x x + λ x x ). Wih he presence of a non-smooh componen, we have he following more general noion. (Gradien mapping) [8] In minimizing f(x) + ψ(x), where f is convex and differeniable and ψ is convex and non-smooh, ( x + = arg min f(x),x x + ) x λ x x + ψ(x) (7) is called he generalized gradien updae, and δ = λ (x x + ) is he (generalized) gradien mapping. Noe ha he quadraic approximaion is made o he smooh componen only. I can be shown ha he gradien mapping is analogous o he gradien in smooh convex opimizaion [4, 8]. This is also a common consruc used in recen sochasic subgradien mehods [3, 7]. 3 Acceleraed Gradien Mehod for Sochasic Learning Le G(x,ξ ) x F(x,ξ ) x=x be he sochasic gradien of F(x,ξ ). We assume ha i is an unbiased esimaor of he gradien f(x), i.e., E ξ [G(x,ξ)] = f(x). Algorihm shows he proposed algorihm, which will be called (Sochasic Acceleraed GradiEn). I involves he updaing of hree sequences {x }, {y } and {z }. Noe ha y is he generalized gradien updae, and x + is a convex combinaion of y and z. The algorihm also mainains wo parameer sequences {α } and {L }. We will see in Secion 3. ha differen seings of hese parameers lead o differen convergence raes. Noe ha he only expensive sep of Algorihm is he compuaion of he generalized gradien updae y, which is analogous o he subgradien compuaion in oher subgradien-based mehods. In general, is compuaional complexiy depends on he srucure of ψ(x). As will be seen in Secion 3.3, his can ofen be efficienly obained in many regularized risk minimizaion problems. Algorihm (Sochasic Acceleraed GradiEn). Inpu: Sequences {L } and {α }. Iniialize: y = z =, α = λ =. L = L + µ. for = o N do x = ( α )y { + α z. y = arg min x G(x,ξ ),x x + L x x + ψ(x) }. z = z (L α + µ) [L (x y ) + µ(z x )]. end for Oupu y N. 3. Convergence Analysis Define G(x,ξ ) f(x ). Because of he unbiasedness of G(x,ξ ), E ξ [ ] =. In he following, we will show ha he value of φ(y ) φ(x) can be relaed o ha of φ(y ) φ(x) for any x. Le δ L (x y ) be he gradien mapping involved in updaing y. Firs, we inroduce he following lemma. Lemma 3. For, φ(x) is quadraically bounded from below as φ(x) φ(y ) + δ,x x + µ x x +,y x + L L L δ. 3

4 Proposiion. Assume ha for each, σ and L > L, hen φ(y ) φ(x) + L α + µα x z ( α )[φ(y ) φ(x)] + L α x z + σ (L L) + α,x z. (8) Proof. Define V (x) = δ,x x + µ x x + Lα x z. I is easy o see ha z = arg min x R d V (x). Moreover, noice ha V (x) is (L α + µ)-srongly convex. Hence on applying Lemmas and 3, we obain ha for any x, V (z ) V (x) L α + µ x z = δ,x x + µ x x + L α x z L α + µ x z φ(x) φ(y ) L L L δ + L α x z L α +µ x z +,x y. Then, φ(y ) can be bounded from above, as: φ(y ) φ(x) + δ,x z L L L δ L α z z + L α x z L (9) α + µ x z +,x y, where he non-posiive erm µ z x has been dropped from is righ-hand-side. On he oher hand, by applying Lemma 3 wih x = y, we ge φ(y ) φ(y ) δ,x y +,y y L L L δ, () where he non-posiive erm µ y x has also been dropped from he righ-hand-side. On muliplying (9) by α and () by α, and hen adding hem ogeher, we obain φ(y ) φ(x) ( α )[φ(y ) φ(x)] L L L δ +A+B+C L α z z, () where A = δ,α (x z )+( α )(x y ), B = α,x y +( α ),y y, and C = Lα x z Lα +µα x z. In he following, we consider o upper bound A and B. Firs, by using he updae rule of x in Algorihm and he Young s inequaliy, we have A = δ,α (x z ) + ( α )(x y ) + α δ,z z = α δ,z z L α z z + δ. () L On he oher hand, B can be bounded as B =,α x + ( α )y x +,x y = α,x z +,δ L α,x z + σ δ, (3) L where he second equaliy is due o he updae rule of x, and he las sep is from he Cauchy- Schwarz inequaliy and he boundedness of. Hence, plugging () and (3) ino (), φ(y ) φ(x) ( α )[φ(y ) φ(x)] (L L) δ L + σ δ + α,x z + C L σ ( α )[φ(y ) φ(x)] + (L L) + α,x z + C, where he las sep is due o he fac ha ax + bx b 4a wih a,b >. On re-arranging erms, we obain (8). The Young s inequaliy saes ha x, y x a + a y for any a >. 4

5 Le he opimal soluion in problem (6) be x. From he updae rules in Algorihm, we observe ha he riple (x,y,z ) depends on he random process ξ [ ] {ξ,...,ξ } and hence is also random. Clearly, z and x are independen of ξ. Thus, E ξ[],x z = E ξ[ ] E ξ[] [,x z ξ [ ] ] = E ξ[ ] E ξ [,x z ] = E ξ[ ] x z, E ξ [ ] =, where he firs equaliy uses E x [h(x)] = E y E x [h(x) y], and he las equaliy is from our assumpion ha he sochasic gradien G(x, ξ) is unbiased. Taking expecaions on boh sides of (8) wih x = x, we obain he following corollary, which will be useful in proving he subsequen heorems. Corollary. E[φ(y )] φ(x ) + L α + µα E[ x z ] ( α )(E[φ(y )] φ(x )) + L α E[ x z ] + σ (L L). So far, he choice of L and α in Algorihm has been lef unspecified. In he following, we will show ha wih a good choice of L and α, (he expecaion of) φ(y ) converges rapidly o φ(x ). Theorem. Assume ha E[ x z ] D for some D. Se L = b( + ) 3 + L, α = +, (4) where b > is a consan. Then he expeced error of Algorihm can be bounded as ) E[φ(y N )] φ(x ) 3D L N + (3D b + 5σ. (5) 3b N If σ were known, we can se b o he opimal choice of 5σ 3D, and he bound in (5) becomes 3D L 5σD N. N + Noe ha so far φ(x) is only assumed o be convex. As is shown in he following heorem, he convergence rae can be furher improved by assuming srong convexiy. This also requires anoher seing of α and L which is differen from ha in (4). Theorem. Assume he same condiions as in Theorem, excep ha φ(x) is µ-srongly convex. Se L = L + µλ, for ; α = λ + λ 4 λ, for, (6) where λ Π k= ( α ) for and λ =. Then, he expeced error of Algorihm can be bounded as E[φ(y N )] φ(x (L + µ)d ) N + 6σ Nµ. (7) In comparison, only converges as O(log(N)/N) for srongly convex objecives. 3. Remarks As in recen sudies on sochasic composie opimizaion [3], he error bounds in (5) and (7) consis of wo erms: a faser erm which is relaed o he smooh componen and a slower erm relaed o he non-smooh componen. benefis from using he srucure of he problem and acceleraes he convergence of he smooh componen. On he oher hand, many sochasic (sub)gradien-based algorihms like do no separae he smooh from he non-smooh par, bu simply rea he whole objecive as non-smooh. Consequenly, convergence of he smooh componen is also slowed down o O(/ N). As can be seen from (5) and (7), he convergence of is essenially encumbered by he variance of he sochasic subgradien. Recall ha he variance of he average of p i.i.d. random 5

6 variables is equal o /p of he original variance. Hence, as in Pegasos [], σ can be reduced by esimaing he subgradien from a daa subse. Unlike he AC-SA algorihm in [3], he seings of L and α in (4) do no require knowledge of σ and he number of ieraions, boh of which can be difficul o esimae in pracice. Moreover, wih he use of a sparsiy-promoing ψ(x), can produce a sparse soluion (as will be experimenally demonsraed in Secion 5) while AC-SA canno. This is because in, he oupu y is obained from a generalized gradien updae. Wih a sparsiy-promoing ψ(x), his reduces o a (sof) hresholding sep, and hus ensures a sparse soluion. On he oher hand, in each ieraion of AC-SA, is oupu is a convex combinaion of wo oher variables. Unforunaely, adding wo vecors is unlikely o produce a sparse vecor. 3.3 Efficien Compuaion of y The compuaional efficiency of Algorihm hinges on he efficien compuaion of y. Recall ha y is jus he generalized gradien updae, and so is no significanly more expensive han he gradien updae in radiional algorihms. Indeed, he generalized gradien updae is ofen a cenral componen in various opimizaion and machine learning algorihms. In paricular, Duchi and Singer [3] showed how his can be efficienly compued wih he various smooh and non-smooh regularizers, including he l,l,l,l, Berhu and marix norms. Ineresed readers are referred o [3] for deails. 4 Acceleraed Gradien Mehod for Online Learning In his secion, we exend he proposed acceleraed gradien scheme for online learning of (). The algorihm, shown in Algorihm, is similar o he sochasic version in Algorihm. Algorihm -based Online Learning Algorihm. Inpus: Sequences {L } and {α }, where L > L and < α <. Iniialize: z = y. loop x = ( α )y + α { z. Oupu y = arg min x f (x ),x x + L x x + ψ(x) }. z = z α (L + µα ) [L (x y ) + µ(z x )]. end loop Firs, we inroduce he following lemma, which plays a similar role as is sochasic counerpar of Lemma 3. Moreover, le δ L (x y ) be he gradien mapping relaed o he updaing of y. Lemma 4. For >, φ (x) can be quadraically bounded from below as φ (x) φ (y ) + δ,x x + µ x x + L L L δ. Proposiion. For any x and, assume ha here exiss a subgradien ĝ(x) ψ(x) such ha f (x) + ĝ(x) Q. Then for Algorihm, φ (y ) φ (x) Q ( α )(L L) + L α x z L + µα α x z + ( α )L α ( α )L y z L z y. Proof Skech. Define τ = L α. From he updae rule of z, one can check ha z = arg min V (x) δ,x x + µ x x x + τ x z. Similar o he analysis in obaining (9), we can obain φ (y ) φ (x) δ,x z L L L δ τ z z + τ x z τ +µ x z. (9) 6 (8)

7 On he oher hand, δ,x z δ L = L ( z x z y ) L z z + L ( α ) z y L α z y, () on using he convexiy of. Using (), he inequaliy (9) becomes φ (y ) φ (x) L ( α ) z y L z y L L L δ + τ x z τ + µ x z. On he oher hand, by he convexiy of φ (x) and he Young s inequaliy, we have φ (y ) φ (y ) f (y ) + ĝ (y ),y y Q () ( α )(L L) + ( α )(L L) y y. () Moreover, by using he updae rule of x and he convexiy of, we have y y = (y x ) + (x y ) = α (y z ) + (x y ) α y z + ( α ) x y = α y z + δ ( α )L. (3) On using (3), i follows from () ha Q φ (y ) φ (y ) ( α )(L L) + α ( α )(L L) y z + L L L δ. Inequaliy (8) hen follows immediaely by adding his o (). Theorem 3. Assume ha µ =, and x z D for. Se α = a and L = al +L, where a (,) is a consan. Then he regre of Algorihm can be bounded as N = [φ (y ) φ (x )] LD a + [ LD + Q ] N. a( a)l Theorem 4. Assume ha µ >, and x z D for. Se α = a, and L = aµ + L + a (µ L) +, where a (,) is a consan. Then he regre of Algorihm can be bounded as N [ (a + a [φ (y ) φ (x )µ + L )] a = ] D + Q log(n + ). a( a)µ In paricular, wih a =, he regre bound reduces o ( 3µ + L) D + Q µ log(n + ). 5 Experimens In his secion, we perform experimens on he sochasic opimizaion of (). Two daa ses are used (Table ). The firs one is he pcmac daa se, which is a subse of he -newsgroup daa se from [8], while he second one is he RCV daa se, which is a filered collecion of he Reuers RCV from [9]. We choose he square loss for l(, ) and he l regularizer for Ω( ) in (). As discussed in Secion 3.3 and [3], he generalized gradien updae can be efficienly compued by sof hresholding in his case. Moreover, we do no use srong convexiy and so µ =. We compare he proposed algorihm (wih L and α in (4)) wih hree recen algorihms: () [3]; () [4]; and (3) [4]. For fair comparison, we compare heir convergence Downloaded from hp://people.cs.uchicago.edu/ vikass/svmlin.hml and hp:// wychen/sc.hml. 7

8 behavior w.r.. boh he number of ieraions and he number of daa access operaions, he laer of which has been advocaed in [4] as an implemenaion-independen measure of ime. Moreover, he efficiency ricks for sparse daa described in [4] are also implemened. Following [4], we se he regularizaion parameer λ in () o 6. The η parameer in is searched over he range of { 6, 5, 4, 3,, }, and he one wih he lowes l -regularized loss is used. As in Pegasos [], he (sub)gradien is compued from small sample subses. The subse size p is se o min(.m,5), where m is he daa se size. This is used on all he algorihms excep, since is based on coordinae descen and is quie differen from he oher sochasic subgradien algorihms. 3 All he algorihms are rained wih he same maximum amoun of ime (i.e., number of daa access operaions). Table : Summary of he daa ses. daa se #feaures #insances sparsiy pcmac 7,5,946.73% RCV 47,36 93,844.% Resuls are shown in Figure. As can be seen, requires much fewer ieraions for convergence han he ohers (Figures (a) and (e)). Moreover, he addiional coss on mainaining x and z are small, and he mos expensive sep in each ieraion is in compuing he generalized gradien updae. Hence, is per-ieraion complexiy is comparable wih he oher (sub)gradien schemes, and is convergence in erms of he number of daa access operaions is sill he fases (Figures (b), (c), (f) and (g)). Moreover, he sparsiy of he soluion is comparable wih hose of he oher algorihms (Figures (d) and (h)). L regularized loss L regularized loss Error (%) Densiy of w Number of Ieraions (a) Number of Daa Accessesx 6 (b) Number of Daa Accessesx 6 (c) Number of Daa Accesses x 6 (d) L regularized loss Number of Ieraions (e) L regularized loss Number of Daa Accessesx 8 (f) Error (%) Number of Daa Accesses x 8 (g) Densiy of w 4 x Number of Daa Accessesx 8 Figure : Performance of he various algorihms on he pcmac (upper) and RCV (below) daa ses. (h) 6 Conclusion In his paper, we developed a novel acceleraed gradien mehod () for sochasic convex composie opimizaion. I enjoys he compuaional simpliciy and scalabiliy of radiional (sub)gradien mehods bu are much faser, boh heoreically and empirically. Experimenal resuls show ha ouperforms recen (sub)gradien descen mehods. Moreover, can also be exended o online learning, obaining he bes regre bounds currenly known. Acknowledgmen This research has been parially suppored by he Research Grans Council of he Hong Kong Special Adminisraive Region under gran For he same reason, an ieraion is also very differen from an ieraion in he oher algorihms. Hence, is no shown in he plos on he regularized loss versus number of ieraions. 8

9 References [] S. Shalev-Shwarz, Y. Singer, and N. Srebro. Pegasos: Primal esimaed sub-gradien solver for SVM. In Proceedings of he 4h Inernaional Conference on Machine Learning, pages 87 84, Corvalis, Oregon, USA, 7. [] A. Bordes, L. Boou, and P. Gallinari. SGD-QN: Careful Quasi-Newon Sochasic Gradien Descen. Journal of Machine Learning Research, : , 9. [3] J. Duchi and Y. Singer. Online and bach learning using forward looking subgradiens. Technical repor, 9. [4] S. Shalev-Shwarz and A. Tewari. Sochasic mehods for l regularized loss minimizaion. In Proceedings of he 6h Inernaional Conference on Machine Learning, pages , Monreal, Quebec, Canada, 9. [5] L. Boou and O. Bousque. The radeoffs of large scale learning. In Advances in Neural Informaion Processing Sysems. 8. [6] S. Shalev-Shwarz and N. Srebro. SVM opimizaion: Inverse dependence on raining se size. In Proceedings of he 5h Inernaional Conference on Machine Learning, pages , Helsinki, Finland, 8. [7] Y. Neserov. A mehod for unconsrained convex minimizaion problem wih he rae of convergence o( k ). Doklady AN SSSR (ranslaed as Sovie. Mah. Docl.), 69: , 983. [8] Y. Neserov. Gradien mehods for minimizing composie objecive funcion. CORE Discussion Paper 7/76, Caholic Universiy of Louvain, Sepember 7. [9] A. Beck and M. Teboulle. A fas ieraive shrinkage-hresholding algorihm for linear inverse problems. SIAM Journal on Imaging Sciences, :83, 9. [] R. Tibshirani. Regression shrinkage and selecion via he Lasso. Journal of he Royal Saisical Sociey: Series B, 58:67 88, 996. [] S. Ji, L. Sun, R. Jin, and J. Ye. Muli-label muliple kernel learning. In Advances in Neural Informaion Processing Sysems. 9. [] S. Ji and J. Ye. An acceleraed gradien mehod for race norm minimizaion. In Proceedings of he Inernaional Conference on Machine Learning. Monreal, Canada, 9. [3] G. Lan. An opimal mehod for sochasic composie opimizaion. Technical repor, School of Indusrial and Sysems Engineering, Georgia Insiue of Technology, 9. [4] Y. Neserov and I.U.E. Neserov. Inroducory Lecures on Convex Opimizaion: A Basic Course. Kluwer, 3. [5] S.M. Kakade and S. Shalev-Shwarz. Mind he dualiy gap: Logarihmic regre algorihms for online opimizaion. In Advances in Neural Informaion Processing Sysems. 9. [6] A. Beck and M. Teboulle. Mirror descen and nonlinear projeced subgradien mehods for convex opimizaion. Operaions Research Leers, 3(3):67 75, 3. [7] S.J. Wrigh, R.D. Nowak, and M.A.T. Figueiredo. Sparse reconsrucion by separable approximaion. In Proceedings of he Inernaional Conference on Acousics, Speech, and Signal Processing, Las Vegas, Nevada, USA, March 8. [8] V. Sindhwani and S.S. Keerhi. Large scale semi-supervised linear SVMs. In Proceedings of he SIGIR Conference on Research and Developmen in Informaion Rerieval, pages , Seale, WA, USA, 6. [9] Y. Song, W.Y. Chen, H. Bai, C.J. Lin, and E.Y. Chang. Parallel specral clusering. In Proceedings of he European Conference on Machine Learning, pages , Anwerp, Belgium, 8. 9

ANALYSIS AND COMPARISONS OF SOME SOLUTION CONCEPTS FOR STOCHASTIC PROGRAMMING PROBLEMS

ANALYSIS AND COMPARISONS OF SOME SOLUTION CONCEPTS FOR STOCHASTIC PROGRAMMING PROBLEMS ANALYSIS AND COMPARISONS OF SOME SOLUTION CONCEPTS FOR STOCHASTIC PROGRAMMING PROBLEMS R. Caballero, E. Cerdá, M. M. Muñoz and L. Rey () Deparmen of Applied Economics (Mahemaics), Universiy of Málaga,

More information

Single-machine Scheduling with Periodic Maintenance and both Preemptive and. Non-preemptive jobs in Remanufacturing System 1

Single-machine Scheduling with Periodic Maintenance and both Preemptive and. Non-preemptive jobs in Remanufacturing System 1 Absrac number: 05-0407 Single-machine Scheduling wih Periodic Mainenance and boh Preempive and Non-preempive jobs in Remanufacuring Sysem Liu Biyu hen Weida (School of Economics and Managemen Souheas Universiy

More information

The Transport Equation

The Transport Equation The Transpor Equaion Consider a fluid, flowing wih velociy, V, in a hin sraigh ube whose cross secion will be denoed by A. Suppose he fluid conains a conaminan whose concenraion a posiion a ime will be

More information

The Application of Multi Shifts and Break Windows in Employees Scheduling

The Application of Multi Shifts and Break Windows in Employees Scheduling The Applicaion of Muli Shifs and Brea Windows in Employees Scheduling Evy Herowai Indusrial Engineering Deparmen, Universiy of Surabaya, Indonesia Absrac. One mehod for increasing company s performance

More information

Large Scale Online Learning.

Large Scale Online Learning. Large Scale Online Learning. Léon Boou NEC Labs America Princeon NJ 08540 leon@boou.org Yann Le Cun NEC Labs America Princeon NJ 08540 yann@lecun.com Absrac We consider siuaions where raining daa is abundan

More information

Distributed and Secure Computation of Convex Programs over a Network of Connected Processors

Distributed and Secure Computation of Convex Programs over a Network of Connected Processors DCDIS CONFERENCE GUELPH, ONTARIO, CANADA, JULY 2005 1 Disribued and Secure Compuaion of Convex Programs over a Newor of Conneced Processors Michael J. Neely Universiy of Souhern California hp://www-rcf.usc.edu/

More information

Analogue and Digital Signal Processing. First Term Third Year CS Engineering By Dr Mukhtiar Ali Unar

Analogue and Digital Signal Processing. First Term Third Year CS Engineering By Dr Mukhtiar Ali Unar Analogue and Digial Signal Processing Firs Term Third Year CS Engineering By Dr Mukhiar Ali Unar Recommended Books Haykin S. and Van Veen B.; Signals and Sysems, John Wiley& Sons Inc. ISBN: 0-7-380-7 Ifeachor

More information

A Note on Using the Svensson procedure to estimate the risk free rate in corporate valuation

A Note on Using the Svensson procedure to estimate the risk free rate in corporate valuation A Noe on Using he Svensson procedure o esimae he risk free rae in corporae valuaion By Sven Arnold, Alexander Lahmann and Bernhard Schwezler Ocober 2011 1. The risk free ineres rae in corporae valuaion

More information

MTH6121 Introduction to Mathematical Finance Lesson 5

MTH6121 Introduction to Mathematical Finance Lesson 5 26 MTH6121 Inroducion o Mahemaical Finance Lesson 5 Conens 2.3 Brownian moion wih drif........................... 27 2.4 Geomeric Brownian moion........................... 28 2.5 Convergence of random

More information

Morningstar Investor Return

Morningstar Investor Return Morningsar Invesor Reurn Morningsar Mehodology Paper Augus 31, 2010 2010 Morningsar, Inc. All righs reserved. The informaion in his documen is he propery of Morningsar, Inc. Reproducion or ranscripion

More information

Mathematics in Pharmacokinetics What and Why (A second attempt to make it clearer)

Mathematics in Pharmacokinetics What and Why (A second attempt to make it clearer) Mahemaics in Pharmacokineics Wha and Why (A second aemp o make i clearer) We have used equaions for concenraion () as a funcion of ime (). We will coninue o use hese equaions since he plasma concenraions

More information

Journal Of Business & Economics Research September 2005 Volume 3, Number 9

Journal Of Business & Economics Research September 2005 Volume 3, Number 9 Opion Pricing And Mone Carlo Simulaions George M. Jabbour, (Email: jabbour@gwu.edu), George Washingon Universiy Yi-Kang Liu, (yikang@gwu.edu), George Washingon Universiy ABSTRACT The advanage of Mone Carlo

More information

Multiprocessor Systems-on-Chips

Multiprocessor Systems-on-Chips Par of: Muliprocessor Sysems-on-Chips Edied by: Ahmed Amine Jerraya and Wayne Wolf Morgan Kaufmann Publishers, 2005 2 Modeling Shared Resources Conex swiching implies overhead. On a processing elemen,

More information

Niche Market or Mass Market?

Niche Market or Mass Market? Niche Marke or Mass Marke? Maxim Ivanov y McMaser Universiy July 2009 Absrac The de niion of a niche or a mass marke is based on he ranking of wo variables: he monopoly price and he produc mean value.

More information

Optimal Stock Selling/Buying Strategy with reference to the Ultimate Average

Optimal Stock Selling/Buying Strategy with reference to the Ultimate Average Opimal Sock Selling/Buying Sraegy wih reference o he Ulimae Average Min Dai Dep of Mah, Naional Universiy of Singapore, Singapore Yifei Zhong Dep of Mah, Naional Universiy of Singapore, Singapore July

More information

E0 370 Statistical Learning Theory Lecture 20 (Nov 17, 2011)

E0 370 Statistical Learning Theory Lecture 20 (Nov 17, 2011) E0 370 Saisical Learning Theory Lecure 0 (ov 7, 0 Online Learning from Expers: Weighed Majoriy and Hedge Lecurer: Shivani Agarwal Scribe: Saradha R Inroducion In his lecure, we will look a he problem of

More information

Stochastic Optimal Control Problem for Life Insurance

Stochastic Optimal Control Problem for Life Insurance Sochasic Opimal Conrol Problem for Life Insurance s. Basukh 1, D. Nyamsuren 2 1 Deparmen of Economics and Economerics, Insiue of Finance and Economics, Ulaanbaaar, Mongolia 2 School of Mahemaics, Mongolian

More information

PROFIT TEST MODELLING IN LIFE ASSURANCE USING SPREADSHEETS PART ONE

PROFIT TEST MODELLING IN LIFE ASSURANCE USING SPREADSHEETS PART ONE Profi Tes Modelling in Life Assurance Using Spreadshees PROFIT TEST MODELLING IN LIFE ASSURANCE USING SPREADSHEETS PART ONE Erik Alm Peer Millingon 2004 Profi Tes Modelling in Life Assurance Using Spreadshees

More information

Why Did the Demand for Cash Decrease Recently in Korea?

Why Did the Demand for Cash Decrease Recently in Korea? Why Did he Demand for Cash Decrease Recenly in Korea? Byoung Hark Yoo Bank of Korea 26. 5 Absrac We explores why cash demand have decreased recenly in Korea. The raio of cash o consumpion fell o 4.7% in

More information

Premium Income of Indian Life Insurance Industry

Premium Income of Indian Life Insurance Industry Premium Income of Indian Life Insurance Indusry A Toal Facor Produciviy Approach Ram Praap Sinha* Subsequen o he passage of he Insurance Regulaory and Developmen Auhoriy (IRDA) Ac, 1999, he life insurance

More information

Bayesian Filtering with Online Gaussian Process Latent Variable Models

Bayesian Filtering with Online Gaussian Process Latent Variable Models Bayesian Filering wih Online Gaussian Process Laen Variable Models Yali Wang Laval Universiy yali.wang.1@ulaval.ca Marcus A. Brubaker TTI Chicago mbrubake@cs.orono.edu Brahim Chaib-draa Laval Universiy

More information

TEMPORAL PATTERN IDENTIFICATION OF TIME SERIES DATA USING PATTERN WAVELETS AND GENETIC ALGORITHMS

TEMPORAL PATTERN IDENTIFICATION OF TIME SERIES DATA USING PATTERN WAVELETS AND GENETIC ALGORITHMS TEMPORAL PATTERN IDENTIFICATION OF TIME SERIES DATA USING PATTERN WAVELETS AND GENETIC ALGORITHMS RICHARD J. POVINELLI AND XIN FENG Deparmen of Elecrical and Compuer Engineering Marquee Universiy, P.O.

More information

Real-time Particle Filters

Real-time Particle Filters Real-ime Paricle Filers Cody Kwok Dieer Fox Marina Meilă Dep. of Compuer Science & Engineering, Dep. of Saisics Universiy of Washingon Seale, WA 9895 ckwok,fox @cs.washingon.edu, mmp@sa.washingon.edu Absrac

More information

Measuring macroeconomic volatility Applications to export revenue data, 1970-2005

Measuring macroeconomic volatility Applications to export revenue data, 1970-2005 FONDATION POUR LES ETUDES ET RERS LE DEVELOPPEMENT INTERNATIONAL Measuring macroeconomic volailiy Applicaions o expor revenue daa, 1970-005 by Joël Cariolle Policy brief no. 47 March 01 The FERDI is a

More information

How To Calculate Price Elasiciy Per Capia Per Capi

How To Calculate Price Elasiciy Per Capia Per Capi Price elasiciy of demand for crude oil: esimaes for 23 counries John C.B. Cooper Absrac This paper uses a muliple regression model derived from an adapaion of Nerlove s parial adjusmen model o esimae boh

More information

How To Predict A Person'S Behavior

How To Predict A Person'S Behavior Informaion Theoreic Approaches for Predicive Models: Resuls and Analysis Monica Dinculescu Supervised by Doina Precup Absrac Learning he inernal represenaion of parially observable environmens has proven

More information

DYNAMIC MODELS FOR VALUATION OF WRONGFUL DEATH PAYMENTS

DYNAMIC MODELS FOR VALUATION OF WRONGFUL DEATH PAYMENTS DYNAMIC MODELS FOR VALUATION OF WRONGFUL DEATH PAYMENTS Hong Mao, Shanghai Second Polyechnic Universiy Krzyszof M. Osaszewski, Illinois Sae Universiy Youyu Zhang, Fudan Universiy ABSTRACT Liigaion, exper

More information

DETERMINISTIC INVENTORY MODEL FOR ITEMS WITH TIME VARYING DEMAND, WEIBULL DISTRIBUTION DETERIORATION AND SHORTAGES KUN-SHAN WU

DETERMINISTIC INVENTORY MODEL FOR ITEMS WITH TIME VARYING DEMAND, WEIBULL DISTRIBUTION DETERIORATION AND SHORTAGES KUN-SHAN WU Yugoslav Journal of Operaions Research 2 (22), Number, 6-7 DEERMINISIC INVENORY MODEL FOR IEMS WIH IME VARYING DEMAND, WEIBULL DISRIBUION DEERIORAION AND SHORAGES KUN-SHAN WU Deparmen of Bussines Adminisraion

More information

Term Structure of Prices of Asian Options

Term Structure of Prices of Asian Options Term Srucure of Prices of Asian Opions Jirô Akahori, Tsuomu Mikami, Kenji Yasuomi and Teruo Yokoa Dep. of Mahemaical Sciences, Risumeikan Universiy 1-1-1 Nojihigashi, Kusasu, Shiga 525-8577, Japan E-mail:

More information

Duration and Convexity ( ) 20 = Bond B has a maturity of 5 years and also has a required rate of return of 10%. Its price is $613.

Duration and Convexity ( ) 20 = Bond B has a maturity of 5 years and also has a required rate of return of 10%. Its price is $613. Graduae School of Business Adminisraion Universiy of Virginia UVA-F-38 Duraion and Convexiy he price of a bond is a funcion of he promised paymens and he marke required rae of reurn. Since he promised

More information

Automatic measurement and detection of GSM interferences

Automatic measurement and detection of GSM interferences Auomaic measuremen and deecion of GSM inerferences Poor speech qualiy and dropped calls in GSM neworks may be caused by inerferences as a resul of high raffic load. The radio nework analyzers from Rohde

More information

Optimal Investment and Consumption Decision of Family with Life Insurance

Optimal Investment and Consumption Decision of Family with Life Insurance Opimal Invesmen and Consumpion Decision of Family wih Life Insurance Minsuk Kwak 1 2 Yong Hyun Shin 3 U Jin Choi 4 6h World Congress of he Bachelier Finance Sociey Torono, Canada June 25, 2010 1 Speaker

More information

SELF-EVALUATION FOR VIDEO TRACKING SYSTEMS

SELF-EVALUATION FOR VIDEO TRACKING SYSTEMS SELF-EVALUATION FOR VIDEO TRACKING SYSTEMS Hao Wu and Qinfen Zheng Cenre for Auomaion Research Dep. of Elecrical and Compuer Engineering Universiy of Maryland, College Park, MD-20742 {wh2003, qinfen}@cfar.umd.edu

More information

Stochastic approximation vis-a-vis online learning for big data analytics

Stochastic approximation vis-a-vis online learning for big data analytics Sochasic approximaion vis-a-vis online learning for big daa analyics Konsaninos Slavakis, Seung-Jun Kim, Gonzalo Maeos, and Georgios B. Giannakis July 30, 2014 We live in an era of daa deluge, where daa

More information

An Online Learning-based Framework for Tracking

An Online Learning-based Framework for Tracking An Online Learning-based Framework for Tracking Kamalika Chaudhuri Compuer Science and Engineering Universiy of California, San Diego La Jolla, CA 9293 Yoav Freund Compuer Science and Engineering Universiy

More information

The option pricing framework

The option pricing framework Chaper 2 The opion pricing framework The opion markes based on swap raes or he LIBOR have become he larges fixed income markes, and caps (floors) and swapions are he mos imporan derivaives wihin hese markes.

More information

A Production-Inventory System with Markovian Capacity and Outsourcing Option

A Production-Inventory System with Markovian Capacity and Outsourcing Option OPERATIONS RESEARCH Vol. 53, No. 2, March April 2005, pp. 328 349 issn 0030-364X eissn 1526-5463 05 5302 0328 informs doi 10.1287/opre.1040.0165 2005 INFORMS A Producion-Invenory Sysem wih Markovian Capaciy

More information

PATHWISE PROPERTIES AND PERFORMANCE BOUNDS FOR A PERISHABLE INVENTORY SYSTEM

PATHWISE PROPERTIES AND PERFORMANCE BOUNDS FOR A PERISHABLE INVENTORY SYSTEM PATHWISE PROPERTIES AND PERFORMANCE BOUNDS FOR A PERISHABLE INVENTORY SYSTEM WILLIAM L. COOPER Deparmen of Mechanical Engineering, Universiy of Minnesoa, 111 Church Sree S.E., Minneapolis, MN 55455 billcoop@me.umn.edu

More information

Online Multi-Class LPBoost

Online Multi-Class LPBoost Online Muli-Class LPBoos Amir Saffari Marin Godec Thomas Pock Chrisian Leisner Hors Bischof Insiue for Compuer Graphics and Vision, Graz Universiy of Technology, Ausria {saffari,godec,pock,leisner,bischof}@icg.ugraz.a

More information

Making a Faster Cryptanalytic Time-Memory Trade-Off

Making a Faster Cryptanalytic Time-Memory Trade-Off Making a Faser Crypanalyic Time-Memory Trade-Off Philippe Oechslin Laboraoire de Securié e de Crypographie (LASEC) Ecole Polyechnique Fédérale de Lausanne Faculé I&C, 1015 Lausanne, Swizerland philippe.oechslin@epfl.ch

More information

Option Put-Call Parity Relations When the Underlying Security Pays Dividends

Option Put-Call Parity Relations When the Underlying Security Pays Dividends Inernaional Journal of Business and conomics, 26, Vol. 5, No. 3, 225-23 Opion Pu-all Pariy Relaions When he Underlying Securiy Pays Dividends Weiyu Guo Deparmen of Finance, Universiy of Nebraska Omaha,

More information

Parameter-Free Convex Learning through Coin Betting

Parameter-Free Convex Learning through Coin Betting JMLR: Workshop and Conference Proceedings 1:1 7, 2016 ICML 2016 AuoML Workshop Parameer-Free Convex Learning hrough Coin Being Francesco Orabona Dávid Pál Yahoo Research, New York FRANCESCO@ORABONA.COM

More information

USE OF EDUCATION TECHNOLOGY IN ENGLISH CLASSES

USE OF EDUCATION TECHNOLOGY IN ENGLISH CLASSES USE OF EDUCATION TECHNOLOGY IN ENGLISH CLASSES Mehme Nuri GÖMLEKSİZ Absrac Using educaion echnology in classes helps eachers realize a beer and more effecive learning. In his sudy 150 English eachers were

More information

Load Prediction Using Hybrid Model for Computational Grid

Load Prediction Using Hybrid Model for Computational Grid Load Predicion Using Hybrid Model for Compuaional Grid Yongwei Wu, Yulai Yuan, Guangwen Yang 3, Weimin Zheng 4 Deparmen of Compuer Science and Technology, Tsinghua Universiy, Beijing 00084, China, 3, 4

More information

Market Liquidity and the Impacts of the Computerized Trading System: Evidence from the Stock Exchange of Thailand

Market Liquidity and the Impacts of the Computerized Trading System: Evidence from the Stock Exchange of Thailand 36 Invesmen Managemen and Financial Innovaions, 4/4 Marke Liquidiy and he Impacs of he Compuerized Trading Sysem: Evidence from he Sock Exchange of Thailand Sorasar Sukcharoensin 1, Pariyada Srisopisawa,

More information

Algorithms for Portfolio Management based on the Newton Method

Algorithms for Portfolio Management based on the Newton Method Algorihms for Porfolio Managemen based on he Newon Mehod Ami Agarwal aagarwal@cs.princeon.edu Elad Hazan ehazan@cs.princeon.edu Sayen Kale sayen@cs.princeon.edu Rober E. Schapire schapire@cs.princeon.edu

More information

Individual Health Insurance April 30, 2008 Pages 167-170

Individual Health Insurance April 30, 2008 Pages 167-170 Individual Healh Insurance April 30, 2008 Pages 167-170 We have received feedback ha his secion of he e is confusing because some of he defined noaion is inconsisen wih comparable life insurance reserve

More information

On the degrees of irreducible factors of higher order Bernoulli polynomials

On the degrees of irreducible factors of higher order Bernoulli polynomials ACTA ARITHMETICA LXII.4 (1992 On he degrees of irreducible facors of higher order Bernoulli polynomials by Arnold Adelberg (Grinnell, Ia. 1. Inroducion. In his paper, we generalize he curren resuls on

More information

cooking trajectory boiling water B (t) microwave 0 2 4 6 8 101214161820 time t (mins)

cooking trajectory boiling water B (t) microwave 0 2 4 6 8 101214161820 time t (mins) Alligaor egg wih calculus We have a large alligaor egg jus ou of he fridge (1 ) which we need o hea o 9. Now here are wo accepable mehods for heaing alligaor eggs, one is o immerse hem in boiling waer

More information

17 Laplace transform. Solving linear ODE with piecewise continuous right hand sides

17 Laplace transform. Solving linear ODE with piecewise continuous right hand sides 7 Laplace ransform. Solving linear ODE wih piecewise coninuous righ hand sides In his lecure I will show how o apply he Laplace ransform o he ODE Ly = f wih piecewise coninuous f. Definiion. A funcion

More information

Particle Filtering for Geometric Active Contours with Application to Tracking Moving and Deforming Objects

Particle Filtering for Geometric Active Contours with Application to Tracking Moving and Deforming Objects Paricle Filering for Geomeric Acive Conours wih Applicaion o Tracking Moving and Deforming Objecs Yogesh Rahi Namraa Vaswani Allen Tannenbaum Anhony Yezzi Georgia Insiue of Technology School of Elecrical

More information

INTEREST RATE FUTURES AND THEIR OPTIONS: SOME PRICING APPROACHES

INTEREST RATE FUTURES AND THEIR OPTIONS: SOME PRICING APPROACHES INTEREST RATE FUTURES AND THEIR OPTIONS: SOME PRICING APPROACHES OPENGAMMA QUANTITATIVE RESEARCH Absrac. Exchange-raded ineres rae fuures and heir opions are described. The fuure opions include hose paying

More information

Principal components of stock market dynamics. Methodology and applications in brief (to be updated ) Andrei Bouzaev, bouzaev@ya.

Principal components of stock market dynamics. Methodology and applications in brief (to be updated ) Andrei Bouzaev, bouzaev@ya. Principal componens of sock marke dynamics Mehodology and applicaions in brief o be updaed Andrei Bouzaev, bouzaev@ya.ru Why principal componens are needed Objecives undersand he evidence of more han one

More information

Vector Autoregressions (VARs): Operational Perspectives

Vector Autoregressions (VARs): Operational Perspectives Vecor Auoregressions (VARs): Operaional Perspecives Primary Source: Sock, James H., and Mark W. Wason, Vecor Auoregressions, Journal of Economic Perspecives, Vol. 15 No. 4 (Fall 2001), 101-115. Macroeconomericians

More information

Performance Center Overview. Performance Center Overview 1

Performance Center Overview. Performance Center Overview 1 Performance Cener Overview Performance Cener Overview 1 ODJFS Performance Cener ce Cener New Performance Cener Model Performance Cener Projec Meeings Performance Cener Execuive Meeings Performance Cener

More information

ARCH 2013.1 Proceedings

ARCH 2013.1 Proceedings Aricle from: ARCH 213.1 Proceedings Augus 1-4, 212 Ghislain Leveille, Emmanuel Hamel A renewal model for medical malpracice Ghislain Léveillé École d acuaria Universié Laval, Québec, Canada 47h ARC Conference

More information

Network Discovery: An Estimation Based Approach

Network Discovery: An Estimation Based Approach Nework Discovery: An Esimaion Based Approach Girish Chowdhary, Magnus Egersed, and Eric N. Johnson Absrac We consider he unaddressed problem of nework discovery, in which, an agen aemps o formulae an esimae

More information

The naive method discussed in Lecture 1 uses the most recent observations to forecast future values. That is, Y ˆ t + 1

The naive method discussed in Lecture 1 uses the most recent observations to forecast future values. That is, Y ˆ t + 1 Business Condiions & Forecasing Exponenial Smoohing LECTURE 2 MOVING AVERAGES AND EXPONENTIAL SMOOTHING OVERVIEW This lecure inroduces ime-series smoohing forecasing mehods. Various models are discussed,

More information

11/6/2013. Chapter 14: Dynamic AD-AS. Introduction. Introduction. Keeping track of time. The model s elements

11/6/2013. Chapter 14: Dynamic AD-AS. Introduction. Introduction. Keeping track of time. The model s elements Inroducion Chaper 14: Dynamic D-S dynamic model of aggregae and aggregae supply gives us more insigh ino how he economy works in he shor run. I is a simplified version of a DSGE model, used in cuing-edge

More information

DOES TRADING VOLUME INFLUENCE GARCH EFFECTS? SOME EVIDENCE FROM THE GREEK MARKET WITH SPECIAL REFERENCE TO BANKING SECTOR

DOES TRADING VOLUME INFLUENCE GARCH EFFECTS? SOME EVIDENCE FROM THE GREEK MARKET WITH SPECIAL REFERENCE TO BANKING SECTOR Invesmen Managemen and Financial Innovaions, Volume 4, Issue 3, 7 33 DOES TRADING VOLUME INFLUENCE GARCH EFFECTS? SOME EVIDENCE FROM THE GREEK MARKET WITH SPECIAL REFERENCE TO BANKING SECTOR Ahanasios

More information

SPEC model selection algorithm for ARCH models: an options pricing evaluation framework

SPEC model selection algorithm for ARCH models: an options pricing evaluation framework Applied Financial Economics Leers, 2008, 4, 419 423 SEC model selecion algorihm for ARCH models: an opions pricing evaluaion framework Savros Degiannakis a, * and Evdokia Xekalaki a,b a Deparmen of Saisics,

More information

Stock Trading with Recurrent Reinforcement Learning (RRL) CS229 Application Project Gabriel Molina, SUID 5055783

Stock Trading with Recurrent Reinforcement Learning (RRL) CS229 Application Project Gabriel Molina, SUID 5055783 Sock raing wih Recurren Reinforcemen Learning (RRL) CS9 Applicaion Projec Gabriel Molina, SUID 555783 I. INRODUCION One relaively new approach o financial raing is o use machine learning algorihms o preic

More information

Online Learning with Sample Path Constraints

Online Learning with Sample Path Constraints Journal of Machine Learning Research 0 (2009) 569-590 Submied 7/08; Revised /09; Published 3/09 Online Learning wih Sample Pah Consrains Shie Mannor Deparmen of Elecrical and Compuer Engineering McGill

More information

Hotel Room Demand Forecasting via Observed Reservation Information

Hotel Room Demand Forecasting via Observed Reservation Information Proceedings of he Asia Pacific Indusrial Engineering & Managemen Sysems Conference 0 V. Kachivichyanuul, H.T. Luong, and R. Piaaso Eds. Hoel Room Demand Forecasing via Observed Reservaion Informaion aragain

More information

Evolutionary building of stock trading experts in real-time systems

Evolutionary building of stock trading experts in real-time systems Evoluionary building of sock rading expers in real-ime sysems Jerzy J. Korczak Universié Louis Paseur Srasbourg, France Email: jjk@dp-info.u-srasbg.fr Absrac: This paper addresses he problem of consrucing

More information

Analysis of Pricing and Efficiency Control Strategy between Internet Retailer and Conventional Retailer

Analysis of Pricing and Efficiency Control Strategy between Internet Retailer and Conventional Retailer Recen Advances in Business Managemen and Markeing Analysis of Pricing and Efficiency Conrol Sraegy beween Inerne Reailer and Convenional Reailer HYUG RAE CHO 1, SUG MOO BAE and JOG HU PARK 3 Deparmen of

More information

Hedging with Forwards and Futures

Hedging with Forwards and Futures Hedging wih orwards and uures Hedging in mos cases is sraighforward. You plan o buy 10,000 barrels of oil in six monhs and you wish o eliminae he price risk. If you ake he buy-side of a forward/fuures

More information

APPLICATION OF THE KALMAN FILTER FOR ESTIMATING CONTINUOUS TIME TERM STRUCTURE MODELS: THE CASE OF UK AND GERMANY. January, 2005

APPLICATION OF THE KALMAN FILTER FOR ESTIMATING CONTINUOUS TIME TERM STRUCTURE MODELS: THE CASE OF UK AND GERMANY. January, 2005 APPLICATION OF THE KALMAN FILTER FOR ESTIMATING CONTINUOUS TIME TERM STRUCTURE MODELS: THE CASE OF UK AND GERMANY Somnah Chaeree* Deparmen of Economics Universiy of Glasgow January, 2005 Absrac The purpose

More information

Dynamic programming models and algorithms for the mutual fund cash balance problem

Dynamic programming models and algorithms for the mutual fund cash balance problem Submied o Managemen Science manuscrip Dynamic programming models and algorihms for he muual fund cash balance problem Juliana Nascimeno Deparmen of Operaions Research and Financial Engineering, Princeon

More information

Life insurance cash flows with policyholder behaviour

Life insurance cash flows with policyholder behaviour Life insurance cash flows wih policyholder behaviour Krisian Buchard,,1 & Thomas Møller, Deparmen of Mahemaical Sciences, Universiy of Copenhagen Universiesparken 5, DK-2100 Copenhagen Ø, Denmark PFA Pension,

More information

Analysis of Planck and the Equilibrium ofantis in Tropical Physics

Analysis of Planck and the Equilibrium ofantis in Tropical Physics Emergence of Fokker-Planck Dynamics wihin a Closed Finie Spin Sysem H. Niemeyer(*), D. Schmidke(*), J. Gemmer(*), K. Michielsen(**), H. de Raed(**) (*)Universiy of Osnabrück, (**) Supercompuing Cener Juelich

More information

Optimal Time to Sell in Real Estate Portfolio Management

Optimal Time to Sell in Real Estate Portfolio Management Opimal ime o Sell in Real Esae Porfolio Managemen Fabrice Barhélémy and Jean-Luc Prigen hema, Universiy of Cergy-Ponoise, Cergy-Ponoise, France E-mails: fabricebarhelemy@u-cergyfr; jean-lucprigen@u-cergyfr

More information

Smooth Priorities for Multi-Product Inventory Control

Smooth Priorities for Multi-Product Inventory Control Smooh rioriies for Muli-roduc Invenory Conrol Francisco José.A.V. Mendonça*. Carlos F. Bispo** *Insiuo Superior Técnico - Universidade Técnica de Lisboa (email:favm@mega.is.ul.p) ** Insiuo de Sisemas e

More information

On Galerkin Approximations for the Zakai Equation with Diffusive and Point Process Observations

On Galerkin Approximations for the Zakai Equation with Diffusive and Point Process Observations On Galerkin Approximaions for he Zakai Equaion wih Diffusive and Poin Process Observaions An der Fakulä für Mahemaik und Informaik der Universiä Leipzig angenommene DISSERTATION zur Erlangung des akademischen

More information

A Distributed Multiple-Target Identity Management Algorithm in Sensor Networks

A Distributed Multiple-Target Identity Management Algorithm in Sensor Networks A Disribued Muliple-Targe Ideniy Managemen Algorihm in Sensor Neworks Inseok Hwang, Kaushik Roy, Hamsa Balakrishnan, and Claire Tomlin Dep. of Aeronauics and Asronauics, Sanford Universiy, CA 94305 Elecrical

More information

adaptive control; stochastic systems; certainty equivalence principle; long-term

adaptive control; stochastic systems; certainty equivalence principle; long-term COMMUICATIOS I IFORMATIO AD SYSTEMS c 2006 Inernaional Press Vol. 6, o. 4, pp. 299-320, 2006 003 ADAPTIVE COTROL OF LIEAR TIME IVARIAT SYSTEMS: THE BET O THE BEST PRICIPLE S. BITTATI AD M. C. CAMPI Absrac.

More information

Chapter 8: Regression with Lagged Explanatory Variables

Chapter 8: Regression with Lagged Explanatory Variables Chaper 8: Regression wih Lagged Explanaory Variables Time series daa: Y for =1,..,T End goal: Regression model relaing a dependen variable o explanaory variables. Wih ime series new issues arise: 1. One

More information

Efficient One-time Signature Schemes for Stream Authentication *

Efficient One-time Signature Schemes for Stream Authentication * JOURNAL OF INFORMATION SCIENCE AND ENGINEERING, 611-64 (006) Efficien One-ime Signaure Schemes for Sream Auhenicaion * YONGSU PARK AND YOOKUN CHO + College of Informaion and Communicaions Hanyang Universiy

More information

On Stochastic and Worst-case Models for Investing

On Stochastic and Worst-case Models for Investing On Sochasic and Wors-case Models for Invesing Elad Hazan IBM Almaden Research Cener 650 Harry Rd, San Jose, CA 9520 ehazan@cs.princeon.edu Sayen Kale Yahoo! Research 430 Grea America Parkway, Sana Clara,

More information

Inventory Planning with Forecast Updates: Approximate Solutions and Cost Error Bounds

Inventory Planning with Forecast Updates: Approximate Solutions and Cost Error Bounds OPERATIONS RESEARCH Vol. 54, No. 6, November December 2006, pp. 1079 1097 issn 0030-364X eissn 1526-5463 06 5406 1079 informs doi 10.1287/opre.1060.0338 2006 INFORMS Invenory Planning wih Forecas Updaes:

More information

Differential Equations and Linear Superposition

Differential Equations and Linear Superposition Differenial Equaions and Linear Superposiion Basic Idea: Provide soluion in closed form Like Inegraion, no general soluions in closed form Order of equaion: highes derivaive in equaion e.g. dy d dy 2 y

More information

DDoS Attacks Detection Model and its Application

DDoS Attacks Detection Model and its Application DDoS Aacks Deecion Model and is Applicaion 1, MUHAI LI, 1 MING LI, XIUYING JIANG 1 School of Informaion Science & Technology Eas China Normal Universiy No. 500, Dong-Chuan Road, Shanghai 0041, PR. China

More information

Research on Inventory Sharing and Pricing Strategy of Multichannel Retailer with Channel Preference in Internet Environment

Research on Inventory Sharing and Pricing Strategy of Multichannel Retailer with Channel Preference in Internet Environment Vol. 7, No. 6 (04), pp. 365-374 hp://dx.doi.org/0.457/ijhi.04.7.6.3 Research on Invenory Sharing and Pricing Sraegy of Mulichannel Reailer wih Channel Preference in Inerne Environmen Hanzong Li College

More information

Option Pricing Under Stochastic Interest Rates

Option Pricing Under Stochastic Interest Rates I.J. Engineering and Manufacuring, 0,3, 8-89 ublished Online June 0 in MECS (hp://www.mecs-press.ne) DOI: 0.585/ijem.0.03. Available online a hp://www.mecs-press.ne/ijem Opion ricing Under Sochasic Ineres

More information

Keldysh Formalism: Non-equilibrium Green s Function

Keldysh Formalism: Non-equilibrium Green s Function Keldysh Formalism: Non-equilibrium Green s Funcion Jinshan Wu Deparmen of Physics & Asronomy, Universiy of Briish Columbia, Vancouver, B.C. Canada, V6T 1Z1 (Daed: November 28, 2005) A review of Non-equilibrium

More information

Sampling Time-Based Sliding Windows in Bounded Space

Sampling Time-Based Sliding Windows in Bounded Space Sampling Time-Based Sliding Windows in Bounded Space Rainer Gemulla Technische Universiä Dresden 01062 Dresden, Germany gemulla@inf.u-dresden.de Wolfgang Lehner Technische Universiä Dresden 01062 Dresden,

More information

Task is a schedulable entity, i.e., a thread

Task is a schedulable entity, i.e., a thread Real-Time Scheduling Sysem Model Task is a schedulable eniy, i.e., a hread Time consrains of periodic ask T: - s: saring poin - e: processing ime of T - d: deadline of T - p: period of T Periodic ask T

More information

Distributed Online Localization in Sensor Networks Using a Moving Target

Distributed Online Localization in Sensor Networks Using a Moving Target Disribued Online Localizaion in Sensor Neworks Using a Moving Targe Aram Galsyan 1, Bhaskar Krishnamachari 2, Krisina Lerman 1, and Sundeep Paem 2 1 Informaion Sciences Insiue 2 Deparmen of Elecrical Engineering-Sysems

More information

A UNIFIED APPROACH TO MATHEMATICAL OPTIMIZATION AND LAGRANGE MULTIPLIER THEORY FOR SCIENTISTS AND ENGINEERS

A UNIFIED APPROACH TO MATHEMATICAL OPTIMIZATION AND LAGRANGE MULTIPLIER THEORY FOR SCIENTISTS AND ENGINEERS A UNIFIED APPROACH TO MATHEMATICAL OPTIMIZATION AND LAGRANGE MULTIPLIER THEORY FOR SCIENTISTS AND ENGINEERS RICHARD A. TAPIA Appendix E: Differeniaion in Absrac Spaces I should be no surprise ha he differeniaion

More information

Strategic Optimization of a Transportation Distribution Network

Strategic Optimization of a Transportation Distribution Network Sraegic Opimizaion of a Transporaion Disribuion Nework K. John Sophabmixay, Sco J. Mason, Manuel D. Rossei Deparmen of Indusrial Engineering Universiy of Arkansas 4207 Bell Engineering Cener Fayeeville,

More information

Optimal Longevity Hedging Strategy for Insurance. Companies Considering Basis Risk. Draft Submission to Longevity 10 Conference

Optimal Longevity Hedging Strategy for Insurance. Companies Considering Basis Risk. Draft Submission to Longevity 10 Conference Opimal Longeviy Hedging Sraegy for Insurance Companies Considering Basis Risk Draf Submission o Longeviy 10 Conference Sharon S. Yang Professor, Deparmen of Finance, Naional Cenral Universiy, Taiwan. E-mail:

More information

Constant Data Length Retrieval for Video Servers with Variable Bit Rate Streams

Constant Data Length Retrieval for Video Servers with Variable Bit Rate Streams IEEE Inernaional Conference on Mulimedia Compuing & Sysems, June 17-3, 1996, in Hiroshima, Japan, p. 151-155 Consan Lengh Rerieval for Video Servers wih Variable Bi Rae Sreams Erns Biersack, Frédéric Thiesse,

More information

arxiv:1309.5977v1 [stat.ml] 23 Sep 2013

arxiv:1309.5977v1 [stat.ml] 23 Sep 2013 EFFICIENT SAMPLING FROM TIME-VARYING LOG-CONCAVE DISTRIBUTIONS By Hariharan Narayanan and Alexander Rakhlin Universiy of Washingon and Universiy of Pennsylvania arxiv:1309.5977v1 [sa.ml] 3 Sep 013 We propose

More information

As widely accepted performance measures in supply chain management practice, frequency-based service

As widely accepted performance measures in supply chain management practice, frequency-based service MANUFACTURING & SERVICE OPERATIONS MANAGEMENT Vol. 6, No., Winer 2004, pp. 53 72 issn 523-464 eissn 526-5498 04 060 0053 informs doi 0.287/msom.030.0029 2004 INFORMS On Measuring Supplier Performance Under

More information

A Generalized Bivariate Ornstein-Uhlenbeck Model for Financial Assets

A Generalized Bivariate Ornstein-Uhlenbeck Model for Financial Assets A Generalized Bivariae Ornsein-Uhlenbeck Model for Financial Asses Romy Krämer, Mahias Richer Technische Universiä Chemniz, Fakulä für Mahemaik, 917 Chemniz, Germany Absrac In his paper, we sudy mahemaical

More information

Forecasting and Information Sharing in Supply Chains Under Quasi-ARMA Demand

Forecasting and Information Sharing in Supply Chains Under Quasi-ARMA Demand Forecasing and Informaion Sharing in Supply Chains Under Quasi-ARMA Demand Avi Giloni, Clifford Hurvich, Sridhar Seshadri July 9, 2009 Absrac In his paper, we revisi he problem of demand propagaion in

More information

UNDERSTANDING THE DEATH BENEFIT SWITCH OPTION IN UNIVERSAL LIFE POLICIES. Nadine Gatzert

UNDERSTANDING THE DEATH BENEFIT SWITCH OPTION IN UNIVERSAL LIFE POLICIES. Nadine Gatzert UNDERSTANDING THE DEATH BENEFIT SWITCH OPTION IN UNIVERSAL LIFE POLICIES Nadine Gazer Conac (has changed since iniial submission): Chair for Insurance Managemen Universiy of Erlangen-Nuremberg Lange Gasse

More information

Differential Equations. Solving for Impulse Response. Linear systems are often described using differential equations.

Differential Equations. Solving for Impulse Response. Linear systems are often described using differential equations. Differenial Equaions Linear sysems are ofen described using differenial equaions. For example: d 2 y d 2 + 5dy + 6y f() d where f() is he inpu o he sysem and y() is he oupu. We know how o solve for y given

More information

Modeling VIX Futures and Pricing VIX Options in the Jump Diusion Modeling

Modeling VIX Futures and Pricing VIX Options in the Jump Diusion Modeling Modeling VIX Fuures and Pricing VIX Opions in he Jump Diusion Modeling Faemeh Aramian Maseruppsas i maemaisk saisik Maser hesis in Mahemaical Saisics Maseruppsas 2014:2 Maemaisk saisik April 2014 www.mah.su.se

More information